[abcnews] fix extraction(closes #12394)(closes #27920)

This commit is contained in:
Remita Amine 2021-01-27 12:28:22 +01:00
parent d18f4419a7
commit 11b68df7a4

View File

@ -1,14 +1,15 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import calendar
import re import re
import time
from .amp import AMPIE from .amp import AMPIE
from .common import InfoExtractor from .common import InfoExtractor
from .youtube import YoutubeIE from ..utils import (
from ..compat import compat_urlparse parse_duration,
parse_iso8601,
try_get,
)
class AbcNewsVideoIE(AMPIE): class AbcNewsVideoIE(AMPIE):
@ -18,8 +19,8 @@ class AbcNewsVideoIE(AMPIE):
(?: (?:
abcnews\.go\.com/ abcnews\.go\.com/
(?: (?:
[^/]+/video/(?P<display_id>[0-9a-z-]+)-| (?:[^/]+/)*video/(?P<display_id>[0-9a-z-]+)-|
video/embed\?.*?\bid= video/(?:embed|itemfeed)\?.*?\bid=
)| )|
fivethirtyeight\.abcnews\.go\.com/video/embed/\d+/ fivethirtyeight\.abcnews\.go\.com/video/embed/\d+/
) )
@ -49,6 +50,12 @@ class AbcNewsVideoIE(AMPIE):
}, { }, {
'url': 'http://abcnews.go.com/2020/video/2020-husband-stands-teacher-jail-student-affairs-26119478', 'url': 'http://abcnews.go.com/2020/video/2020-husband-stands-teacher-jail-student-affairs-26119478',
'only_matching': True, 'only_matching': True,
}, {
'url': 'http://abcnews.go.com/video/itemfeed?id=46979033',
'only_matching': True,
}, {
'url': 'https://abcnews.go.com/GMA/News/video/history-christmas-story-67894761',
'only_matching': True,
}] }]
def _real_extract(self, url): def _real_extract(self, url):
@ -69,28 +76,23 @@ class AbcNewsIE(InfoExtractor):
_VALID_URL = r'https?://abcnews\.go\.com/(?:[^/]+/)+(?P<display_id>[0-9a-z-]+)/story\?id=(?P<id>\d+)' _VALID_URL = r'https?://abcnews\.go\.com/(?:[^/]+/)+(?P<display_id>[0-9a-z-]+)/story\?id=(?P<id>\d+)'
_TESTS = [{ _TESTS = [{
'url': 'http://abcnews.go.com/Blotter/News/dramatic-video-rare-death-job-america/story?id=10498713#.UIhwosWHLjY', # Youtube Embeds
'url': 'https://abcnews.go.com/Entertainment/peter-billingsley-child-actor-christmas-story-hollywood-power/story?id=51286501',
'info_dict': { 'info_dict': {
'id': '10505354', 'id': '51286501',
'ext': 'flv', 'title': "Peter Billingsley: From child actor in 'A Christmas Story' to Hollywood power player",
'display_id': 'dramatic-video-rare-death-job-america', 'description': 'Billingsley went from a child actor to Hollywood power player.',
'title': 'Occupational Hazards',
'description': 'Nightline investigates the dangers that lurk at various jobs.',
'thumbnail': r're:^https?://.*\.jpg$',
'upload_date': '20100428',
'timestamp': 1272412800,
}, },
'add_ie': ['AbcNewsVideo'], 'playlist_count': 5,
}, { }, {
'url': 'http://abcnews.go.com/Entertainment/justin-timberlake-performs-stop-feeling-eurovision-2016/story?id=39125818', 'url': 'http://abcnews.go.com/Entertainment/justin-timberlake-performs-stop-feeling-eurovision-2016/story?id=39125818',
'info_dict': { 'info_dict': {
'id': '38897857', 'id': '38897857',
'ext': 'mp4', 'ext': 'mp4',
'display_id': 'justin-timberlake-performs-stop-feeling-eurovision-2016',
'title': 'Justin Timberlake Drops Hints For Secret Single', 'title': 'Justin Timberlake Drops Hints For Secret Single',
'description': 'Lara Spencer reports the buzziest stories of the day in "GMA" Pop News.', 'description': 'Lara Spencer reports the buzziest stories of the day in "GMA" Pop News.',
'upload_date': '20160515', 'upload_date': '20160505',
'timestamp': 1463329500, 'timestamp': 1462442280,
}, },
'params': { 'params': {
# m3u8 download # m3u8 download
@ -102,49 +104,55 @@ class AbcNewsIE(InfoExtractor):
}, { }, {
'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343', 'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343',
'only_matching': True, 'only_matching': True,
}, {
# inline.type == 'video'
'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343',
'only_matching': True,
}] }]
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) story_id = self._match_id(url)
display_id = mobj.group('display_id') webpage = self._download_webpage(url, story_id)
video_id = mobj.group('id') story = self._parse_json(self._search_regex(
r"window\['__abcnews__'\]\s*=\s*({.+?});",
webpage, 'data'), story_id)['page']['content']['story']['everscroll'][0]
article_contents = story.get('articleContents') or {}
webpage = self._download_webpage(url, video_id) def entries():
video_url = self._search_regex( featured_video = story.get('featuredVideo') or {}
r'window\.abcnvideo\.url\s*=\s*"([^"]+)"', webpage, 'video URL') feed = try_get(featured_video, lambda x: x['video']['feed'])
full_video_url = compat_urlparse.urljoin(url, video_url) if feed:
yield {
'_type': 'url',
'id': featured_video.get('id'),
'title': featured_video.get('name'),
'url': feed,
'thumbnail': featured_video.get('images'),
'description': featured_video.get('description'),
'timestamp': parse_iso8601(featured_video.get('uploadDate')),
'duration': parse_duration(featured_video.get('duration')),
'ie_key': AbcNewsVideoIE.ie_key(),
}
youtube_url = YoutubeIE._extract_url(webpage) for inline in (article_contents.get('inlines') or []):
inline_type = inline.get('type')
if inline_type == 'iframe':
iframe_url = try_get(inline, lambda x: x['attrs']['src'])
if iframe_url:
yield self.url_result(iframe_url)
elif inline_type == 'video':
video_id = inline.get('id')
if video_id:
yield {
'_type': 'url',
'id': video_id,
'url': 'http://abcnews.go.com/video/embed?id=' + video_id,
'thumbnail': inline.get('imgSrc') or inline.get('imgDefault'),
'description': inline.get('description'),
'duration': parse_duration(inline.get('duration')),
'ie_key': AbcNewsVideoIE.ie_key(),
}
timestamp = None return self.playlist_result(
date_str = self._html_search_regex( entries(), story_id, article_contents.get('headline'),
r'<span[^>]+class="timestamp">([^<]+)</span>', article_contents.get('subHead'))
webpage, 'timestamp', fatal=False)
if date_str:
tz_offset = 0
if date_str.endswith(' ET'): # Eastern Time
tz_offset = -5
date_str = date_str[:-3]
date_formats = ['%b. %d, %Y', '%b %d, %Y, %I:%M %p']
for date_format in date_formats:
try:
timestamp = calendar.timegm(time.strptime(date_str.strip(), date_format))
except ValueError:
continue
if timestamp is not None:
timestamp -= tz_offset * 3600
entry = {
'_type': 'url_transparent',
'ie_key': AbcNewsVideoIE.ie_key(),
'url': full_video_url,
'id': video_id,
'display_id': display_id,
'timestamp': timestamp,
}
if youtube_url:
entries = [entry, self.url_result(youtube_url, ie=YoutubeIE.ie_key())]
return self.playlist_result(entries)
return entry